Palm Vein Recognition Based on Dual-channel Convolutional Neural Network

Palm vein recognition technology is an emerging biometric technology, which has a wide application prospect in the security field. It has the advantages of liveness detection, uniqueness and unforgeability. The accuracy of current palm vein recognition algorithm is not high enough, especially when processing the non-contact palm vein image. To solve this problem, a palm vein recognition based on dual-channel convolutional neural network (D-CNN) is proposed in this paper. In therecognition algorithm, the raw images are inputted into the first channel to acquire the whole characteristics of the image. And the palm vein image been positioned and histogram equalized are inputted into the second channel to obtain the enhanced local detail vein characteristics. Simulation results show that, D-CNN can effectively extract the palm vein features with higher accuracy than current methods.